After your AutoML tabular classification model is done training, create an endpoint and deploy your model to the endpoint. After your model is deployed to this new endpoint, test your model by requesting a prediction.
This tutorial has several pages:
Deploying the model to an endpoint and sending a prediction.
Each page assumes that you have already performed the instructions from the previous pages of the tutorial.
Deploy your model to an endpoint
When your model finishes training, it is listed in the Models tab.
In the Google Cloud Console, in the Vertex AI section, go to the Models page.
Find your model, and click its link to open its Evaluate panel.
This panel displays quality metrics for the model, including a confusion matrix. You can select a value for the target column to see evaluation metrics for that value. Below, you can see how strongly each column affected model training (Feature importance).
Open the Deploy & Test panel. Under Deploy your model, click Deploy to endpoint.
Structured_AutoML_Tutorialfor the endpoint name and select the model you just created.
Accept the defaults for traffic split, minimum and maximum number of compute nodes, and the machine type.
Click Continue, then click Deploy to create your endpoint and deploy your model to it.
Deploying a model can take several minutes.
Request a prediction
While the endpoint is being created, you can optionally enter a set of values for a prediction. Return to the Models list in the left-hand navigation panel and open your newly created model.
Open the Deploy & test tab.
You can use the prefilled values for the prediction data or enter your own.
When the model is deployed, click Predict.
For this model, a prediction result of
1represents a negative outcome—a deposit is not made at the bank. A prediction result of
2represents a positive outcome—a deposit is made at the bank.
If you used the pre-filled prediction values, the local feature importance values are all zero. This is because the pre-filled values are the baseline prediction data, so the prediction returned is the baseline prediction value.
Follow the last page of the tutorial to clean up resources that you have created.